Aldehyde dehydrogenase as a potential target for toxicant- induced Parkinson’s disease
4. Discussion
pesticide may be associated with increased PD risk and dopaminergic cell damage independent of α-synuclein levels.
rather than residential addresses. Analyses of ambient exposures based on residential addresses showed no association with PD, likely because the population was randomly distributed at residential addresses during evenings or weekends when commercial pesticides were not applied. Notably, this model only estimates relative ambient exposures and does not take into account other routes of exposures such as occupational handling of pesticides and drinking well water, both associated with increased risk of PD, which could potentiate the estimated risk even further.88,133,134
Our finding that ambient benomyl exposure was associated with increased PD risk offered an excellent opportunity to investigate the hypothesis that this association is causal and to determine the underlying mechanisms of toxicity. It is very difficult to establish causality from an epidemiologic association, particularly for a risk-increasing factor such as a toxicant. One approach to this end is to determine if exposure in experimental models can recapitulate some of the pathologic features of the disease.
Since concentrations in the range of 1.3-50 µM are necessary for benomyl’s intended use as a fungicide,135-137 we used a conservative concentration range of 0.1-1 µM to ensure relevance to actual exposures for humans. Here we report relatively selective dopaminergic neuronal damage in both an in vitro and a novel in vivo model. We found significant losses of TH+ neurons in primary cultures exposed to benomyl at low
concentrations, while the survival of other neurons was unaffected. In order to further evaluate benomyl’s selective toxicity, we developed a novel in vivo zebafish model.
Zebrafish are vertebrates with a well-developed dopaminergic system, and since the embryos are transparent, we can visualize specific neuronal types using fluorescent reporter genes. Two different transgenic lines were used: ETvmat2:GFP zebrafish124 in
which the aminergic neurons are labeled with GFP, and Tg(sensory:GFP) zebrafish125 that label trigeminal and Rohon-Beard sensory neurons. Aminergic neurons in zebrafish larvae were selectively damaged while sensory neurons were preserved after benomyl exposure, in a similar manner as observed in the mesencephalic cultures. In analyzing aminergic neurons, we focused on the most prominent clusters at 5 days
postfertilization. The anterior clusters shown in Figure III-3a correspond to aminergic neurons in the olfactory bulb and telencephalon; the posterior clusters are diencephalic.
Holzschuh et al. previously reported that these clusters also include (nor)adrenergic neurons, although they are predominantly dopaminergic.138 There was significant loss in these clusters, whereas the non-aminergic trigeminal and Rohon-Beard neurons did not undergo any damage, suggesting selectivity for dopaminergic neurons.
Benomyl’s action as a fungicide is thought to be due to its ability to impair microtubule assembly, interfering with mitosis and other processes.139,140 We found previously that microtubule inhibitors such as nocodazole and rotenone also inhibit the UPS.62 Here we report that the carbendazim moiety of benomyl, a known microtubule inhibitor, confers benomyl’s UPS inhibitory activity. It is unlikely that microtubule or UPS dysfunction is responsible for benomyl’s toxicity observed in these primary cultures or zebrafish, given the low concentrations used. The IC50 for UPS inhibition by benomyl is 5.7 µM, which is more than an order of magnitude higher than the concentrations needed to damage dopaminergic neurons in primary cultures. Furthermore,
carbendazim had no effect on neuronal survival at 1 µM, whereas MBT recapitulated benomyl’s toxicity. Our data show that benomyl and MBT, but not carbendazim, inhibit ALDH activity in primary neurons. Since MBT is the toxic metabolite, ALDH inhibition
likely confers benomyl’s neurotoxicity. Wey et al. recently reported TH+ neuronal loss in Aldh1a1-/-xAldh2-/- mice.141 Although environmental investigations like ours can be confounded by multiple simultaneous toxic mechanisms, their genetic approach
targeting ALDH1 and ALDH2 yielded a result similar to what we observed in vitro and in vivo, supporting our hypothesis that benomyl exposure damages dopaminergic neurons via ALDH inhibition. The present work adds an epidemiologic PD association to
demonstrate the relevance of this mechanism to PD pathogenesis.
Our observed alterations in DA metabolites after benomyl exposure add further support for ALDH inhibition as the cause of benomyl’s neurotoxicity. Wey et al. also reported DOPAL accumulation concomitant with TH+ neuronal loss and behavioral deficits in mice lacking cytosolic and mitochondrial ALDHs.141 Burke et al. reported that DOPAL is 400-fold more toxic to dopaminergic neurons in vivo than its precursor (DA) or its ALDH product (DOPAC).14,142,143 Thus, ALDH inhibition leads to the accumulation of DOPAL and provides a mechanism for benomyl’s selective toxicity to dopaminergic neurons. Consistent with this hypothesis, reducing DOPAL synthesis with an MAO inhibitor mitigated benomyl’s toxicity.
In summary, pesticide exposure has been suggested to be a major risk factor for PD, and here we report an epidemiologic association between chronic benomyl
exposure and increased PD risk. We found that dopaminergic neurons are selectively vulnerable to benomyl exposure both in neuronal cultures and in a novel in vivo
zebrafish model. Benomyl and its metabolites are potent ALDH inhibitors, and several converging lines of evidence reveal ALDH inhibition as a mechanism of toxicity by which benomyl and other environmental toxicants may be associated with PD risk.
Augmenting ALDH activity may be a new therapeutic target for designing disease- modifying therapies for PD.
Figure III-1. Aldehyde dehydrogenase inhibition as a potential mechanism of benomyl- induced Parkinson’s disease. Benomyl is efficiently metabolized to several potent ALDH inhibitors, so exposure leads to the accumulation of the toxic dopamine metabolite DOPAL. This offers a possible explanation for the selective toxicity to dopaminergic neurons observed in PD pathogenesis. ALDH is a potential therapeutic target to alter disease progression.
Abbreviations: BIC, butyl isocyanate; MBT, S-methyl N-butylthiocarbamate; MBT-SO, S- methyl N-butylthiocarbamate sulfoxide; DA, dopamine; DOPAL, 3,4-
dihydroxyphenylacetaldehyde; DOPAC, 3,4-dihydroxyphenylacetic acid; GSH, glutathione; CYP, cytochrome P450; MAO, monoamine oxidase; ALDH, aldehyde dehydrogenase; Cys, cysteine residue
Figure III-2. Dopaminergic neuronal damage in primary mesencephalic cultures exposed to benomyl or its metabolites. Representative field of view (20x) shows immunoreactive (a) dopaminergic neurons (TH+) and (b) neuronal nuclei (NeuN+). (c) Exposure to 100 nM or 1 mM benomyl results in 24±6% (P=0.0017, n=16) and 35±4%
(P=5.5x10-10, n=47) respective losses of TH+ neurons as compared to untreated cultures. MBT exposure recapitulates this effect (P=5.5x10-5, n=28), whereas carbendazim does not significantly damage dopaminergic neurons (P=0.73, n=10).
Since MBT is either the proximal or penultimate benomyl metabolite that inhibits ALDH activity, there appears to be an association between neuronal damage and ALDH inhibition; proteasomal inhibition by the carbendazim moiety is not sufficient to kill cells under the same conditions. *P<0.05, **P<0.01, ***P<0.0001
Figure III-3. Aminergic neuronal damage in Danio rerio larvae exposed to benomyl.
Representative confocal images of zebrafish larvae (a,c,e) unexposed or (b,d,f) bathed in 1 µM benomyl from 5 hours postfertilization until 5 days postfertilization are shown.
(g) Neuronal counts (a-b) decreased in VMAT2+ anterior and diencephalic clusters of ETvmat2:GFP zebrafish exposed to benomyl (solid bars) but were unaffected in (c-d) Rohon-Beard and (e-f) trigeminal neurons in Tg(sensory:GFP) zebrafish. (h)
Measurement of total fluorescence yielded the same results. *P<0.1, **P<0.05 Abbreviations: LC, locus coeruleus; De, diencephalon; OB, olfactory bulb; Te, telencephalon
Figure III-4. Inhibitory actions of benomyl and its metabolites. (a) Exposure to benomyl or MBT inhibited ALDH activity ex vivo in mesencephalic neurons dissociated from 2- day-old rat pups (n=3-11). (b) Benomyl inhibited in vitro ALDH activity in rat hepatic mitochondria preparations with an IC50 of 140±19 nM (n=2-4). BIC had essentially the same effect (IC50=120±32 nM; n=3-4); MBT was somewhat less potent with an IC50 of 1.3±0.2 µM (n=4-8). Carbendazim did not significantly inhibit ALDH activity at up to 20 µM (n=4). (c) A reporter protein revealed that benomyl inhibited 26S UPS activity in SK-
N-MC neuroblastoma cells with an IC50 of 5.7±0.5 µM (n=4-14). Carbendazim exposure had the same effect (IC50 of 5.7±0.3 µM, n=4-14), whereas MBT exposure up to 10 µM had no effect (n=5). (d) Benomyl and its BIC/MBT metabolites inhibit ALDH activity at lower concentrations than benomyl and its carbendazim moiety inhibit the 26S UPS.
*P<0.01, **P<0.0001
Abbreviations: SE, standard error; 26S UPS, ubiquitin-proteasome system
Figure III-5. Neuroprotection via reducing DOPAL accumulation with MAO inhibitor. The neuronal loss resulting from 1 µM benomyl or MBT exposure was mitigated by co-
treatment with the MAO inhibitor pargyline (200 µM, n=13-28). Because MAO-B
inhibition reduces the metabolism of dopamine to DOPAL, this suggests that DOPAL is toxic to dopaminergic neurons, and that benomyl is toxic via DOPAL accumulation as a result of ALDH inhibition. *P=0.0027, **P=2.4x10-4, ***P=6.1x10-5
Table III-1. Associations between PD and estimated ambient occupational or residential benomyl exposures
Exposurea
Cases (n=360)
Controls
(n=754) ORb (95%CI) P-valuec Occupational, n(%)
No 217 (60.3) 531 (70.4) 1.00
Low 49 (13.6) 106 (14.1) 1.06 (0.71-1.59) 0.78
High 94 (26.1) 117 (15.5) 1.65 (1.17-2.32) 0.004 Residential, n(%)
No 209 (58.1) 467 (61.9) 1.00
Low 72 (20.0) 143 (19.0) 1.01 (0.70-1.45) 0.96
High 79 (21.9) 144 (19.1) 1.00 (0.70-1.43) 0.99
Abbreviations: OR = unconditional logistic odds ratio; CI = confidence interval
a “Low” exposure defined as below the median value in exposed controls; “High”
exposure defined as equal to or above the median value in exposed controls.
b adjusted for age (continuous), sex (male/female), smoking status (current, former, never), county (Fresno, Kern, Tulare), and education (<12 yrs, =12 yrs, >12 yrs).
c For multiple testing considerations, six tests were performed and a p-value of 0.008 was considered statistically significant.
Supplementary Table III-1. Demographics of the Parkinson’s Environment & Genes (PEG) Study
Parameter
Cases (n=360)
Controls
(n=754) P-valuea Gender, n (%)
Female 154 (42.8) 402 (53.3)
Male 206 (57.2) 352 (46.7) 0.001
Ageb
Mean (sd) 68.3 (10.2) 66.9 (11.2) 0.036
Range 34-88 35-99
Racec, n (%)
Caucasian, non-Hispanic 290 (80.6) 526 (69.8)
Caucasian, Hispanic 47 (13.1) 144 (19.1)
Native American 16 (4.4) 35 (4.6)
African American 3 (0.8) 25 (3.3)
Asian American 4 (1.1) 22 (2.9) 0.002
County, n (%)
Fresno 163 (45.3) 310 (41.1)
Kern 129 (35.8) 322 (42.7)
Tulare 68 (18.9) 122 (16.2) 0.087
Smoking, n (%)
Never 188 (52.2) 364 (48.3)
Former 152 (42.2) 213 (28.3)
Current 20 (5.6) 177 (23.5) <0.001
Education, n (%)
< 12 years 67 (18.6) 111 (14.7)
= 12 years 96 (26.7) 156 (20.7)
> 12 years 197 (54.7) 487 (64.6) 0.007
Family history of PDd, n (%)
No 307 (85.3) 691 (91.6)
Yes 53 (14.7) 63 (8.4) 0.001
a P-value for chi-square test or t-test
b age at diagnosis for cases; age at interview for controls
c race information unrecorded for two controls
d defined as first-degree relative with PD; data unrecorded and assumed negative for 27 controls
Supplementary Table III-2. Associations between PD risk and estimated ambient occupational or residential benomyl exposures, by quartiles in exposed controls
Exposurea
Cases (n=360)
Controls
(n=754) ORb (95%CI) P-valuec Occupational, n(%)
No 217 (60.3) 531 (70.4) 1.00
1st Quartile 23 (6.4) 55 (7.3) 0.98 (0.57-1.68) 0.93 2nd Quartile 26 (7.2) 51 (6.8) 1.14 (0.67-1.94) 0.62 3rd Quartile 40 (11.1) 62 (8.2) 1.38 (0.87-2.18) 0.18 4th Quartile 54 (15.0) 55 (7.3) 1.92 (1.25-2.96) 0.003
p-value for trend 0.0029 Residential, n(%)
No 209 (58.1) 467 (61.9) 1.00
1st Quartile 31 (8.6) 71 (9.4) 0.86 (0.53-1.39) 0.53 2nd Quartile 41 (11.4) 72 (9.5) 1.17 (0.75-1.85) 0.49 3rd Quartile 31 (8.6) 73 (9.7) 0.83 (0.51-1.34) 0.45 4th Quartile 48 (13.3) 71 (9.4) 1.17 (0.75-1.81) 0.50 p-value for trend 0.72
Abbreviations: OR = unconditional logistic odds ratio; CI = confidence interval
a Exposure quartiles defined using distributions of exposed controls, separately for occupational or residential estimates
b adjusted for age (continuous), sex (male/female), smoking status (current, former, never), county (Fresno, Kern, Tulare), and education (<12 yrs, =12 yrs, >12 yrs).
c For multiple testing considerations, six tests were performed so a P-value of 0.008 was considered statistically significant.
Supplementary Table III-3. Associations between PD risk and estimated ambient occupational or residential benomyl exposures, stratified by sex
Males Females
Exposurea
Cases (n=206)
Controls
(n=352) ORb (95%CI) P-valuec
Cases (n=154)
Controls
(n=402) ORb (95%CI) P-valuec Occupational, n(%)
No 112 (54.4) 235 (66.8) 1.00 105 (68.2) 296 (73.6) 1.00
Low 30 (14.6) 54 (15.3) 1.16
(0.68-2.00) 0.59 19 (12.3) 52 (12.9) 0.91
(0.49-1.69) 0.76
High 64 (31.1) 63 (17.9) 1.81
(1.15-2.84) 0.010 30 (19.5) 54 (13.4) 1.42
(0.83-2.43) 0.20 Residential, n(%)
No 114 (55.3) 216 (61.4) 1.00 95 (61.7) 251 (62.4) 1.00
Low 44 (21.4) 61 (17.3) 1.24
(0.75-2.02) 0.40 28 (18.2) 82 (20.4) 0.80
(0.46-1.38) 0.42
High 48 (23.3) 75 (21.3) 1.06
(0.65-1.71) 0.82 31 (20.1) 69 (17.2) 0.92
(0.53-1.58) 0.76 Abbreviations: OR = unconditional logistic odds ratio; CI = confidence interval
a “Low” exposure defined as below the median value in exposed controls; “High” exposure defined as equal to or above the median value in exposed controls.
b adjusted for age (continuous), sex (male/female), smoking status (current, former, never), county (Fresno, Kern, Tulare), and education (<12 yrs, =12 yrs, >12 yrs).
c For multiple testing considerations, six tests were performed and a p-value of 0.008 was considered statistically significant.